innsight: Get the Insights of your Neural Network

Interpretability methods to analyze the behavior and individual predictions of modern neural networks. Implemented methods are: 'Connection Weights' described by Olden et al. (2004) <doi:10.1016/j.ecolmodel.2004.03.013>, Layer-wise Relevance Propagation ('LRP') described by Bach et al. (2015) <doi:10.1371/journal.pone.0130140>, Deep Learning Important Features ('DeepLIFT') described by Shrikumar et al. (2017) <arXiv:1704.02685> and gradient-based methods like 'SmoothGrad' described by Smilkov et al. (2017) <arXiv:1706.03825>, 'Gradient x Input' described by Baehrens et al. (2009) <arXiv:0912.1128> or 'Vanilla Gradient'.

Version: 0.2.0
Depends: R (≥ 3.5.0)
Imports: checkmate, cli, ggplot2, methods, R6, torch
Suggests: covr, GGally, grid, gridExtra, gtable, keras, knitr, luz, neuralnet, palmerpenguins, plotly, rmarkdown, spelling, tensorflow, testthat (≥ 3.0.0)
Published: 2023-04-16
Author: Niklas Koenen ORCID iD [aut, cre], Raphael Baudeu [ctb]
Maintainer: Niklas Koenen <niklas.koenen at>
License: MIT + file LICENSE
NeedsCompilation: no
Language: en-US
Materials: README NEWS
CRAN checks: innsight results


Reference manual: innsight.pdf
Vignettes: Example 1: Iris dataset with torch
Example 2: Penguin dataset with torch and luz
In-depth explanation
Introduction to innsight


Package source: innsight_0.2.0.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): innsight_0.2.0.tgz, r-oldrel (arm64): innsight_0.2.0.tgz, r-release (x86_64): innsight_0.2.0.tgz, r-oldrel (x86_64): innsight_0.2.0.tgz
Old sources: innsight archive


Please use the canonical form to link to this page.